Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

From his booth at the greasy spoon diner, Mark watched the world go by through a haze of apathy. The clatter of plates, the chatter of strangers – they all felt muted, distant. He was stuck, a broken record repeating the same monotonous tune, just like his life. Then, a song began playing on the jukebox: his song, "Dust and Decay."

The singer had a powerful voice, but the soulfulness of the original was gone, replaced by a generic pop sheen. He winced. The joy he once felt in creating it had shriveled to nothing, and his song was now just another background noise, a commodity. He sighed, pushed back from the table, and walked out, seeking a place where he could finally feel something real again.

Emotion: gloomy

Cluster: Sadness / Despair
PC1 (Valence): -1.68 Negative
PC2 (Disposition): -1.37

Role in Research

This story is one of 1,000 stories generated for the emotion gloomy. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 gloomy stories, after denoising with neutral dialogue baselines, produces the gloomy emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the gloomy vector is projected through the unembedding matrix.

Promoted:
😞0.368
无法0.319
😔0.300
いつも0.270
쓸0.255
Suppressed:
la-0.432
l-0.419
de-0.398
B-0.294
!-0.259